¹Syracuse Research Corporation, North Syracuse, NY, USA; ²Agency for Toxic Substances and Disease Registry, Atlanta, GA, USA.
Sponsor: P. McGinnis

Abstract
This health consultation was prepared to support assessment of the public health implications of leakage of large amounts of CCA from defective shipping containers to soil. Conclusions are applicable to other spills and releases of this chemical mixture. CCA, a wood-preserving pesticide, contains chromium(VI), arsenic(V), and copper(II) in an aqueous solution or concentrate. Concerns for long-term exposure to CCA included persistence in soil, potential runoff with consequent contamination of shellfish and fish, and leaching to sources of drinking water. Evaluation of experimental studies of CCA fate in soil and monitoring studies of wood-preserving facilities where CCA was spilled on the soil indicated that chromium(VI), arsenic, and copper can leach from soil into groundwater and surface water. Leaching potential decreased in the order chromium(VI)>arsenic>copper. CCA persisted in soil and remained leachable for years. Chromium(VI) is not likely to accumulate in fish and shellfish. Copper may accumulate in tissues of mussels and oysters. Arsenic may accumulate, primarily in a relatively nontoxic form, in tissues of fish and shellfish. Evaluation of health effects studies on CCA indicated little or no harm to workers exposed at low levels, but data were inadequate for derivation of a health guideline. Adverse effects in people and animals from high-level exposure were similar to those from the individual components of CCA. These effects included irritant or corrosive effects (all three components), dermal effects (arsenic), neurological effects [arsenic and chromium(VI)], hematological and renal effects (all three). Additional effects of concern for oral exposure to CCA include hepatic effects (all three components) and cancer (arsenic). A weight-of-evidence analysis predicted that interactions among the components are likely to be less than additive or additive. Thus, no increase in toxicity beyond that expected from additive toxicity towards shared target organs is expected for CAA.

Cadmium Pharmacokinetics/Pharmacodynamics (PKPD) Model for Use in Risk Assessment

G.L. Diamond¹, H. Choudhury², and W.C. Thayer¹

¹Syracuse Research Corporation, Syracuse, NY; ²U.S. EPA, National Center for Environmental Assessment, Cincinnati, OH

Abstract
Risks of kidney toxicity in humans exposed to cadmium from the ingestion pathway can be estimated using a PKPD model. The model was used to interconvert external and internal cadmium dose estimates from epidemiological studies, allowing direct comparison of dose-response relationships across studies that might otherwise be incomparable. Based on estimates from 15 epidemiological studies of diverse exposures (occupational, general environmental, environmental contamination), the median kidney cadmium level associated with a 0.1 probability (K₁₀) of low molecular weight proteinuria was predicted to be 108 μg Cd/g renal cortex (95% confidence limits: 70‑240). The model predicted that the lower confidence limit on the K₁₀ (K_10L, 70 μg Cd/g cortex) would be reached at age 55 years (the age at which kidney cadmium levels are predicted to peak) with a constant chronic intake of 1.1 μg/kg/day in females or 2.5 μg/kg/day in males. A lower cadmium intake in females is predicted to result in a similar kidney cadmium level as in males because females were assumed to absorb a larger fraction (10% vs 5%) of an ingested cadmium dose (Choudhury et al., 2001). The K_10L exceeded the peak kidney cadmium level (at age 55 years) predicted to result from the average dietary intake of cadmium in U.S. non-smokers (28 μg Cd/g cortex in female, 14 μg Cd/g cortex in males, and also exceeded the 95^th percentile kidney cadmium levels predicted to result from dietary intakes in the U.S. (43 μg Cd/g cortex in females or 23 μg/g cortex in males). That these values are lower than the K_10L, suggests that, for most of the U.S. population, risks of kidney toxicity from dietary intake of cadmium are negligible. (Views expressed in this report are those of the authors and do not necessarily represent the views or policies of the U.S. EPA.)

Provisional Toxicity Values for 0-Nitrotoluene for the Health Effects Assessment Summary Table (HEAST)

Abstract
EPA is updating toxicity values for chemicals listed in the 1997 HEAST. The previous assessment for o-nitrotoluene (o-NT) included a subchronic RfD of 1E-1 mg/kg-day and a chronic RfD of 1E-2 mg/kg-day, based on a LOAEL of 200 mg/kg for splenic lesions in a 6-month rat gavage study. Subchronic and chronic NTP feeding bioassays in rodents provide a basis for deriving new toxicity values for o-NT. In the subchronic assay, a LOAEL of 44 mg/kg-day was identified for increased levels of methemoglobin in female rats after 3 weeks; this LOAEL was considered minimal because other erythrocytic effects (anemia, splenic lesions) were limited to the higher doses. A provisional subchronic RfD of 4E-2 mg/kg-day was derived from the minimum LOAEL. A significant finding in the subchronic assay was the occurrence of malignant mesotheliomas in the epididymal tunica vaginalis in rats exposed at 353 mg/kg-day for 13 weeks. Similar mesotheliomas and a hepatic cholangiocarcinoma were observed in rats fed 292 mg/kg-day for 13 or 26 weeks in another assay. No provisional chronic RfD was derived because frank effects were observed at the lowest tested dose in the chronic feeding assay (25 mg/kg-day). In the chronic assays, clear evidence for multiple-organ carcinogenicity was observed: subcutaneous skin fibroma/fibrosarcoma in male and female rats, mammary fibroadenoma in female rats, hemangiosarcoma in male and female mice, and hepatocellular adenoma/carcinoma in female mice. o-NT is a genotoxic carcinogen following bioactivation; direct evidence for mutagenicity in cell regulatory genes (p53, K-ras, and -catenin) was shown in mice exposed for 2 years. Dose-response modeling was performed based on the most prominent tumor types. The provisional oral slope factor of 2.3E-1 (mg/kg-day)^-1 was based on the incidence of skin tumors in male rats, the tumor type associated with the highest numerical risk. (This abstract does not necessarily reflect EPA policy.)

ATSDR'S Acute Minimal Risk Level for Copper

L. Ingerman¹ and A.S. Dorsey²

¹Syracuse Research Corporation, Saratoga Springs, NY;
²ATSDR, Atlanta GA
Sponsor: P McGinnis

Abstract
ATSDR is currently re-evaluating the noncancer toxicity of copper and has prepared a draft for public comment version of the toxicological profile for copper. One of the most commonly reported adverse health effects of copper is gastrointestinal distress. Most of the available data on gastrointestinal effects comes from case reports of individuals ingesting beverages stored in copper or untinned brass containers or first draw water. Vomiting, nausea, and/or abdominal pain are usually reported shortly after ingestion. These symptoms are not usually persistent or associated with other effects. The case reports provide limited information on exposure levels, however, recently conducted studies identify a threshold for gastrointestinal effects. These studies, as well as the case reports, provide a strong basis for an acute-duration oral minimal risk level (MRL) for copper. An acute MRL is an estimate of daily exposure to a substance that is likely to be without an appreciable risk of adverse effects over a period of 14 days or less. In studies involving a single exposure to copper, adverse gastrointestinal effects (nausea, vomiting, abdominal pain, and/or diarrhea) have been observed at copper concentrations of 4 mg/L and higher (≥0.01 mg Cu/kg). In repeated exposure studies, vomiting, nausea, and/or abdominal pain have been observed at 5 mg Cu/L (0.073 mg Cu/kg/day); no adverse effects were observed at concentrations of 1 or 3 mg/L. Animal studies support the identification of the gastrointestinal tract as a sensitive target of toxicity. A 2-week human exposure study (Pizarro et al. 1999) was selected as the basis for the acute MRL. No adverse effects were observed in women ingesting 3 mg Cu/L (0.027 mg Cu/kg/day). To estimate total copper exposure, the copper in the drinking water was added to the reported average dietary copper intake of 0.0266 mg Cu/kg/day. The total copper dose of 0.0538 mg/kg/day was considered a NOAEL for gastrointestinal effects. The NOAEL was divided by an uncertainty factor of 3 (to account for human variability) to yield an acute-duration oral MRL of 0.02 mg Cu/kg/day.

Risks to Children from Exposure to Lead in Air During Remedial or Removal Activities at Superfund Sites:
A Case Study of the RSR Lead Smelter Superfund Site

G.A. Khoury¹ and G.L. Diamond²

¹U.S. EPA, Region 6, Dallas, TX;
²Syracuse Research Corporation, Syracuse, NY

Abstract
This study explored modeling approaches for assessing potential risks to children from air lead emissions during removal of a smelter facility. The EPA Integrated Exposure Uptake Biokinetic (IEUBK) model and the International Commission of Radiologic Protection (ICRP) lead model were used to simulate blood lead concentrations (PbB) in children, based on monitored air lead concentrations. Although air lead concentrations at monitoring stations located in the downwind community intermittently exceeded the NAAQ standard for lead (1.5 μg/m³), both models indicated that exposures to children in the community areas did not pose a significant long-term or acute risk. Long-term risk was defined as greater than 5% probability of a child having a long-term PbB that exceeded 10 μg/dL, the CDC and the EPA concern level. Short-term or acute risk was defined as greater than 5% probability of a child having a PbB on any given day that exceeded 20 μg/dL, the CDC trigger level for medical evaluation. The models were also used to estimate air lead levels for short- or long-term exposures that would not exceed specified levels of risk (risk-based concentrations, RBCs) for given daily exposure durations (3 or 8 hr/day) or frequencies (1–7 days/week). RBCs for long-term exposures, based on the ICRP model, ranged from 0.3 (7 days/week, 8 hr/day) to 4.4 μg/m³ (1 day/week, 3 hr/day) and were lower than those based on the IEUBK model (1.9–18 μg/m³). RBCs short-term exposures based on the ICRP model ranged from 3.5 to 29.0 μg/m³. Recontamination of remediated residential yards from deposition of air lead emitted during remedial activities at the site was also examined. The increase in soil lead concentration due to lead deposition in the community at large was predicted to be 3.0 mg/kg; this would be less than 1% increase when compared to the cleanup level of 500 mg/kg developed for residential yards at the site. (Views are those of the authors and not necessarily of any Federal agency or department.)

IDLH Documentation Review

R. McCartney¹, L. Jackson¹, P. McGinnis¹, A. Maier², H.A. Ahlers³

¹Syracuse Research Corporation, Cincinnati, OH; ²Toxicology Excellence for Risk Assessment, Cincinnati, OH; ³National Institute for Occupational Safety and Health, Cincinnati, OH USA

Abstract
Immediately Dangerous to Life and Health (IDLH) values have been established by the National Institute for Occupational Safety and Health (NIOSH) to protect workers against exposure conditions that may cause severe irreversible health effects or impair escape from exposure environments. Original assessments were used primarily as a tool to recommend respiratory protection and IDLH values were often based on limited scientific data, secondary source reported data, or safety considerations. In 1994, NIOSH reevaluated the IDLH values and developed more formal criteria for determining IDLH values. A review of the toxicological basis for 35 of a total 398 of the current IDLH values was conducted to determine if the existing IDLH recommendations were consistent with current toxicological data. For this critical analysis, a qualitative rating method was developed as a tool for conducting the evaluation. Technical guidelines were established for selection of experimental studies, evaluation of each study, and evaluation of the database for each substance. Experimental study protocol and methods were recorded into a customized database designed for tracking and sorting scientific data specific to acute exposures to hazardous substances. Two overall evaluations were made for each substance. The first evaluation rated the adequacy of the database to support the development of an IDLH value and classified each as adequate, marginally adequate, or inadequate. Out of a total 35 substance databases, approximately 50% were considered marginally adequate and 50% adequate. The second evaluation rated the degree to which the current IDLH value was likely to be protective against dangerous levels of exposure. Of the current 35 IDLH values reviewed, approximately 20% were considered more than adequately protective, 63% were considered protective, and 17% were considered inadequately protective.

New Approaches for Deriving a Reference Concentration for
Methyl Ethyl Ketone

M.H. Follansbee, P.R. McClure, and P. McGinnis

Syracuse Research Corporation, North Syracuse, NY, USA

Abstract
MEK is listed on IRIS with a reference concentration (RfC) of 1 mg/m³ based on developmental effects (decreased fetal body weight and increased incidence of misaligned sternebrae) reported in CD-1 mice exposed 7 h/day on gestation days 6-15 (Schwetz et al., 1991). This derivation used a NOAEL/LOAEL approach. A review of the available toxicity data for methyl ethyl ketone (MEK) uncovered no new data for inhalation exposures. Current EPA guidance provides for a number of alternative approaches to derive an RfC. A benchmark dose approach to derivation of an RfC for MEK was explored using the same data for developmental effects in CD-1 mice, as well as data for increased incidence of extra ribs in Sprague-Dawley rats as reported by an inhalation developmental study (Deacon et al., 1981). Modeling of these data and calculation of benchmark concentrations (BMC) and 95% confidence limits on the BMC (BMCL) identified points of departure for each endpoint. The points of departure for each endpoint fall between the NOAEL and LOAEL observed in the respective developmental studies; thus, all are equally plausible. The available pharmacokinetic, pharmacodynamic, and mechanism of action data do not provide sufficient evidence to argue convincingly for either peak exposure level or area under the curve as the most appropriate surrogate metric for internal effective dose; consequently, duration adjustment was also explored in order to time-weight the exposures used in the critical studies for each endpoint. In addition, an adjustment can be made to account for the difference in the blood:air absorption coefficient between rats and humans based on literature reports. This exercise identifies and explores a number of alternative approaches to derive an RfC for MEK, which primarily reflects new approaches to RfC derivation, rather than new data. The impact of each of these decisions on the RfC will be presented.